The sensitivity of an ion analytical instrument, such as a mass spectrometer, depends in part upon the efficiency with which the coupled ion source generates ions from the analytical sample and then delivers those ions to the instrument for analysis. Matrix Assisted Laser Desorption and Ionization (MALDI)/Laser Desorption Ionization (LDI) and Electrospray ion sources have become an essential and enabling building block in modern mass spectrometry of biological macromolecules (e.g. proteins, peptides and sugars etc.). Both methods were awarded the Nobel Prize in Chemistry in 2002 and revolutionized the application of mass spectrometers in life science, in particular in proteomics but also in functional genomics and metabolomics and drug discovery.
One of the ultimate goals of life science (including disciplines such as proteomics) is the prediction of disease based on molecular information. To achieve this goal, highly efficient and sensitive ion sources as well as mass spectrometers with sufficient mass accuracy and reliability have to be available. Highly sensitive ion sources are needed in view of factors such as the following:                1. The human proteome is estimated to contain >106 protein species.        2. These proteins are thought to occur at an extremely large range of abundance (≈1010).        3. Typically purification/selective binding are required which further reduces sample abundance.        4. Many purification methods impact ionization efficiencies negatively.        5. Frequently only very small sample amounts in low concentrations are available (mg or less, a few 103 cells).        6. Investigations face combinatorial complexity from factors such as the very high number of measurements required, and the fact that Investigations limited by time.        
It can be assumed that the discovery of biomarkers and the ability to predict diseases is currently hindered and limited by the unreliability with which distinctive patterns in mass spectra can be found, at least partially as a result of imperfections and limitation of the current ion source technology. This is one of the remaining obstacles for mass spectrometry to move from being an instrument in biochemical labs to an everyday tool in clinics and hospitals. Beside the application in life science and medicine, rapid and sensitive detection of organic and inorganic compounds will, unfortunately, become more common in the form of screening for biological agents and residues of explosives.
One specific variant of MALDI is sometimes referred to as Surface Enhanced Laser Desorption Ionization (SELDI) in which the matrix is already pre-deposited on the target surface. We will henceforth refer to MALDI and SELDI commonly as MALDI, or in general, as LDI. Electrospray (ES) has a number derivatives or of what can be considered variants such as Electrohydrodynamic ionization, Aerospray ionization, ACPI, and Thermospray ionization and which shall also be considered as included in the following.
Fundamentally, both LDI and ES methods suffer from a number of problems which limit their practical application, including aspects such as sensitivity, usefulness as quantitative tools, and usefulness in biomarker discovery.
One significant problematic factor is that of molecular fragmentation. Due to the high laser power densities LDI/MALDI ion sources eject ions with substantial translational and internal temperatures which frequently results in molecular fragmentation and decay thereby limiting the available ion life time for analysis. Such ion fragmentation also reduces sensitivity and, importantly, reduces the ‘fidelity’ or clarity of mass spectra which limits or prevents further data analysis, e.g. for biomarker discovery and analysis: Correlation of data from mass spectra with medical conditions of living or dead, human, animal or plant subjects from which analyzed samples were taken.
In MALDI ion sources typically a UV laser (sometimes IR) is fired at the crystals in the MALDI spot with typical pulse duration on the order of tLP≈10−9 to 10−8 s. The matrix molecules in the spot absorb the electromagnetic laser energy and are thought to protect the sample molecules. This, however, is only achieved to a very limited extent.
Originally, LDI/MALDI ion sources have been operated under vacuum conditions at pressures where sample ion-background gas collisions are negligible. Later, ion sources operating at elevated pressure or Atmospheric Pressure MALDI (AP MALDI) have been introduced for convenience in terms of sample handling as well as collisional cooling.
Experiments carried out in the early 90's indicated improved ion transmission within gas-filled multipole ion guides due to “collisional cooling”: Repeated collisions of ions with gas molecules reduce the temperature of the ions and also cause the ion beam to collapse axially inside RF multipole ion guides.
This collisional cooling effect was subsequently utilized in MALDI ion sources themselves. Simple versions of so called elevated pressure and Atmospheric Pressure MALDI (AP MALDI) ion sources have been described beginning in the late '90. However, their ion-optical design is poor and a pneumatic design is effectively non-existent due to the lack of appropriate computational design tools capable of modeling the flow field as well as the electro-pneumatic interactions (ion-neutral collisions).
A second significant problematic factor in conventional mass spectrometry involves inefficiency. LDI/MALDI is a highly inefficient means to generate ions from a sample which results in a general lack of sensitivity of this method as well as very poor performance in terms of true quantitative sample analysis. In addition, sample preparation techniques strongly influence the characteristics of the obtained mass-spectrometric data in a mostly unpredictable manner.
In LDI/MALDI ion sources molecules in the sample spot absorb the electromagnetic laser energy and it is thought that primarily the matrix is ionized by this event. The matrix is then thought to transfer part of their charge to the analyte (e.g. a protein), thus ionizing them while (to a limited extent) still protecting them from the disruptive energy of the laser. Ions observed after this process are quasimolecular ions that are typically ionized by the addition of a proton to [M+H]+ or the removal of a proton [M−H]−. MALDI generally produces singly-charged ions, but multiply-charged ions such as [M+2H]2+ have been observed specifically in conjunction with IR lasers.
However, if one thoroughly analyzes the budget of ions in a mass spectrometer it becomes apparent that the total ionization efficiency of MALDI is incredibly weak. For example, if a sample of 1 pmol (6·1023·10−12=6·1011) of stable biological macromolecules with a mass on the order of m=103 u is introduced into a commercially available high-end MALDI triple-quadrupole-Time-of-Flight (TOF) instrument an ion count on the order 104 can be expected. It is known that the total ion transmission efficiency of that particular type of mass spectrometer (including detector efficiency, duty cycle, quadrupole transmission etc.) is on the order of 10−2. This means that approximately only 106 ions are transmitted from the MALDI ion source into the mass spectrometer.
Since the sample contains 6·1011 molecules the ionization efficiency is on the order of 106/6·1011≈1.6·10−6. Approximately only one sample molecule in one million becomes an ion and is transmitted into the mass spectrometer. Even if this approximation would underestimate the ionization efficiency by one order of magnitude it is still apparent that a fundamental shortcoming of state-of-the-art MALDI is the lack of ionization efficiency. Further improvements in mass spectrometer performance can be helpful but have by far less potential than improvements on the ion sources and aspects such as ionization efficiency.
In conventional MALDI ion sources the available time for ionization is approximately only on the order of the duration of the laser pulse or slightly above (t≈101 ns). Thereafter, the plume expands and electrons and protons are rapidly extracted from the plume due their substantially lower mass-to-charge ratio m/q compared to sample ions of interest with a typical range of m/q≈102 u/e to 106 u/e.
The creation and transfer of free charges to sample molecules in a conventional MALDI process can in fact be considered a byproduct.
A third level of problematic considerations involve the electrospray process itself. In Electrospray ion sources a liquid, in which the sample molecules are dissolved, is pressed through a capillary. It is generally assumed that the sample molecules are already in an ionized state inside the liquid and upon leaving the capillary the liquid forms a mist (or aerosol) of very small droplets containing such ionized sample molecules (“nebulization”) which, due to coulombic forces, eventually releases individual ionized sample molecules of varying charge state. The exact mechanism of the ion formation is a matter of scientific debate.
There are several fundamental problems in Electrospray ion sources. First, the nebulization and ionization depends on large number of parameters such as, sample concentration, degree of dissociation, liquid flow rate, liquid conductivity, liquid surface tension, capillary diameter, liquid pressure, electric field, gas flow fields, gas temperature fields, gas pressure fields, etc. Stable nebulization and ionization can be difficult to achieve. Moreover, a single or a plurality of droplets can not intentionally be created at a specific point in time with specific initial velocity and direction. Further The total ionization efficiency is also very low (although generally assumed to be better than conventional MALDI) since it depends to some extend on physical characteristics of the initial droplets and their creation, such as net charge, which are at least partially influenced by or in fact based on random processes/natural fluctuation.
A fourth problem is that thus far ES ion source designs have been considerably suboptimal since the combined influence of the electric fields and gas flow fields has not been addressed with sufficient accuracy due to the lack of appropriate computational tools.
A fifth level of problematic considerations involve operational limitations. Both advanced LDI and ES ion sources are inherently difficult to operate due to the complexity of the ion source behavior, the number of parameters that can be adjusted, and the limited available time during measurements. A typical user of such ion sources (connected to mass spectrometers) can not be expected to perform such correcting adjustments in an optimal and rapid fashion.
Thus there is a need to address this problem by providing an automated, active control and feedback system which performs the desired operations.